Multi-source information fusion road surface recognition and tire active heating control system

Abstract

The invention discloses a pavement recognition and tire active heating control system based on multi-source information fusion, and relates to the technical field of vehicle safety control. The system comprises a multi-source information fusion pavement recognition module, a tire temperature rise control strategy module and an active tire heating control module which are in signal connection in sequence; the multi-source information fusion pavement recognition module is used for recognizing pavement types and evaluating pavement key parameters; the tire temperature rise control strategy module is used for generating independent tire target temperatures of four wheels; the active tire heating control module is used for regulating and controlling the tire temperature to the target temperature. According to the method, the target temperature is calculated by fusing multi-modal sensor data, combining an improved SVM classification model and a scene adaptation rule and utilizing double paths (the road surface type and the slip rate), a confidence coefficient weighting and dynamic arbitration mechanism is introduced, and the road surface recognition accuracy and the temperature control precision are improved.

Description

Technical Field

[0001] This invention belongs to the field of vehicle safety control technology, and in particular relates to a multi-source information fusion road surface recognition and active tire heating control system, which is applicable to driving safety control of various vehicles such as cars, SUVs, and commercial vehicles under low adhesion and complex road surface conditions. Background Technology

[0002] When a vehicle is driving on different road conditions (such as dry ground, wet ground, snow, and ice), the road surface adhesion directly affects the tire grip, which in turn affects driving stability and safety. Especially on low-temperature, low-traction roads, the low tire operating temperature will cause the rubber hardness to increase, the grip to decrease significantly, and may easily lead to safety hazards such as increased braking distance and loss of steering control.

[0003] In the prior art, there are some solutions for vehicle energy control. For example, patent CN110752653A discloses a vehicle and its energy control method and device, which provides power to the tire assist device through a battery. However, this solution does not specify a specific road surface recognition scheme, nor does it propose a closed-loop control strategy for wheel temperature. It cannot accurately adjust the tire temperature according to different road conditions, making it difficult to meet the vehicle's handling and stability requirements under complex road conditions.

[0004] Therefore, there is an urgent need for a control system that can accurately identify road surface types and dynamically optimize tire temperature. By fusing multi-source information, it can accurately judge road conditions and, combined with a scientific heating control strategy, maintain tire temperature within the optimal range, thereby improving driving safety on low-traction roads. Summary of the Invention

[0005] The purpose of this invention is to provide a road surface recognition and active tire heating control system based on multi-source information fusion. By fusing data from multiple sensors, it achieves accurate road surface recognition. Combined with a hierarchical heating control strategy and a closed-loop heating execution mechanism, it enables on-demand, zoned, and precise control of tire temperature, solving the problem that existing vehicles cannot accurately adjust tire temperature according to different road conditions.

[0006] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution:

[0007] This invention relates to a multi-source information fusion road surface recognition and active tire heating control system, comprising a multi-source information fusion road surface recognition module, a tire heating control strategy module, and an active tire heating control module connected in sequence by signals. The multi-source information fusion road surface recognition module is used to identify road surface types and evaluate key road surface parameters. The tire heating control strategy module is used to generate independent target temperatures for four tires. The active tire heating control module is used to regulate the tire temperature to the target temperature. The modules are connected in sequence by signals to collaboratively achieve road surface recognition and precise tire heating control.

[0008] As a preferred technical solution, the multi-source information fusion road surface recognition module includes an information input unit, an information preprocessing unit, a feature extraction unit, a model classification and decision-making unit, and a recognition result output unit;

[0009] The information input unit includes a visible light camera, lidar, infrared temperature sensor, and road surface estimation algorithm built into the vehicle, used to collect road surface images, distance, temperature, and vehicle driving-related data.

[0010] The information preprocessing unit is used to perform time synchronization (aligning to a unified timestamp) and Gaussian filtering noise reduction on the raw data from multiple sensors. Outlier handling criteria and Min-Max feature standardization are used to form spatiotemporally consistent feature vectors.

[0011] The feature extraction unit extracts road surface color histogram features, texture LBP features, geometric contour features, and vehicle driving state (wheel speed, acceleration) related features from preprocessed data through a learning model, and outputs a low-dimensional feature vector.

[0012] The model classification and decision-making unit adopts an improved support vector machine classification model to process low-dimensional feature vectors and output the probability distribution of dry land, wet land, snow, ice, curves, and split roads. The decision-making module outputs the final road recognition result based on the maximum probability principle and scene adaptation rules.

[0013] The identification result output unit transmits the road surface type, estimated friction coefficient, and road surface temperature to the tire heating control strategy module.

[0014] As a preferred technical solution, the tire temperature rise control strategy module includes a road surface information classification unit, a temperature rise estimation model and arbitration unit, and a complex road surface temperature rise arbitration unit;

[0015] The road surface information classification unit divides the road surface into a single state (such as pure dry ground or pure snow) or a composite state (such as snow and curves) through logical judgment.

[0016] The temperature rise estimation model and arbitration unit include an information input and calculation subunit, an arbitration and decision-making subunit, and a result output subunit.

[0017] The complex road surface temperature rise arbitration unit determines whether the road surface is complex or sloping by using road slope sensors and steering angle sensors; when it detects the interpolation of the friction coefficients of the left and right road surfaces... At that time, the complex road surface is a split road surface, and the target temperature of the tire on the low friction coefficient side will be increased. When the steering angle The complex road surface is a curved road surface. The roll moment is calculated based on the steering angle, and the target temperature of the tire on the load-off side increases. When the slope The complex road surface is a curved road surface, and the target temperature of the drive wheels increases when going uphill. The target temperature of the brake wheel increases when going downhill. The final output includes four independent target temperatures. The These correspond to the front left, front right, rear left, and rear right tires, respectively; among them, The values ​​were obtained from tables based on the temperatures of complex road surfaces.

[0018] As a preferred technical solution, the information input and calculation subunit calculates the target temperature through a dual-path approach;

[0019] Path 1: Introducing road surface information confidence level ; Value The calculation is based on the number of occurrences of various road surface results and the scene level (security level 1-5) within 10 recognition periods, weighted by the average. ;

[0020] The target temperature 1 is calculated by querying the baseline temperature corresponding to each road surface condition through a preset road surface type-temperature mapping table; the specific formula is as follows:

[0021] ;

[0022] In the formula, For the target temperature, For the first Number of times this type of road surface occurs For the first Safety level of road surface;

[0023] Path Two: Based on the vehicle's actual speed, wheel angular velocity, and effective wheel rolling radius collected by the vehicle body sensors, the slip ratio formula is used. Calculate the initial slip ratio;

[0024] Introducing sensor information confidence (Value) The slip ratio is estimated by weighting the proportion of samples below freezing point among 20 samples collected by an infrared temperature sensor within 5 seconds (based on signal stability). The slip ratio is estimated by classifying and weighting the number of samples collected from the infrared road surface temperature signal. The final estimated slip ratio is obtained; the target temperature is obtained by querying the preset slip ratio-target temperature mapping table. .

[0025] As a preferred technical solution, the arbitration and decision-making subunit acquires the target temperature. Target temperature In addition to tire temperature rise characteristics, a temperature interpolation threshold is set. ;when When, priority arbitration shall be executed; when At that time, the first period is based on weight. Calculate the initial target temperature Subsequent cycles will be based on the temperature growth rate. , Adjust the weights; if and But currently Previous cycle Then keep Use the current target temperature; otherwise, output the normal temperature value.

[0026] As a preferred technical solution, the active tire heating control module includes a temperature acquisition unit, a heating control unit, a current drive unit, and a tire-embedded heating element; the heating control unit is pre-calibrated for the tire at different initial temperatures (- ), ambient temperature ( The drive current required to heat up to the target temperature is used as feedforward compensation with the corresponding PWM duty cycle (resolution 1%). Combined with closed-loop control, the output is dynamically adjusted, and the heating element is controlled by the current drive unit to achieve closed-loop heating of the tire.

[0027] As a preferred technical solution, in the model classification and decision unit, the classification model analyzes the feature vectors and outputs the probability distribution of various road surfaces, and the decision module comprehensively derives the final road surface recognition result based on the probability distribution.

[0028] As a preferred technical solution, the heating element is attached to the inner side of the tire tread or integrated into the tire rubber composite material to achieve zoned heating of the tire contact area; the active tire heating control module controls the heating process based on the difference between the actual temperature and the target temperature.

[0029] As a preferred technical solution, the system further includes a collaborative pre-aiming unit with the vehicle navigation system and the front vision sensor; the collaborative pre-aiming unit is configured to: identify ahead curves, slopes or potential low-traction road surfaces based on navigation path information and visual pre-aiming information, and instruct the tire heating control module to start or adjust the heating strategy in advance.

[0030] As a preferred technical solution, the system's workflow is as follows:

[0031] Step S1: Continuously collect road surface data and vehicle dynamic data through multi-source sensors;

[0032] Step S2: Identify the current road surface type and estimate its state parameters based on the collected data;

[0033] Step S3: Determine the target heating parameters for at least one tire based on the identified road surface type and condition parameters;

[0034] Step S4: Based on the determined target heating parameters, generate a control signal to the heating actuator;

[0035] Step S5: Adjust the heating power in real time based on tire temperature feedback.

[0036] The present invention has the following beneficial effects:

[0037] (1) This invention integrates multimodal sensor data, combines an improved SVM classification model with scene adaptation rules, uses dual paths (road surface type + slip ratio) to calculate the target temperature, introduces confidence weighting and dynamic arbitration mechanism, improves the road surface recognition accuracy and temperature control precision, and avoids the temperature fluctuation problem of the existing "on-off" heating.

[0038] (2) The present invention designs a zoned temperature rise compensation strategy for road surfaces with splits, turns and slopes, taking into account the differences in axle load distribution, which significantly improves the vehicle's handling stability under complex working conditions and enhances its adaptability to complex road surfaces.

[0039] (3) This invention adopts zoned heating elements and feedforward + closed-loop control, which increases the heating speed by 40% (it only takes 3 minutes to heat from -10℃ to 14℃), reduces energy consumption during the heat preservation stage by 30%, and balances heating effect and economy. It has temperature deviation protection, overcurrent / overtemperature protection functions, IP67 waterproof rating for heating elements, and a service life of ≥5000h, making it suitable for... A wide temperature range working environment.

[0040] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0041] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a schematic diagram of a multi-source information fusion road surface recognition and active tire heating control system according to the present invention.

[0043] Figure 2 A flowchart for road surface information recognition based on multi-source information fusion;

[0044] Figure 3 Flowchart of the tire temperature control strategy for the control system;

[0045] Figure 4 Flowchart of the control strategy for the temperature rise estimation model and arbitration module;

[0046] Figure 5 A flowchart for the arbitration process of temperature rise on complex road surfaces;

[0047] Figure 6 Flowchart of tire temperature control strategy. Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0050] To make the purpose, technical solution, and advantages of this application clearer, the following description is provided in conjunction with the appendix. Figure 1-6 The present application will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the application.

[0051] The multi-source information fusion road surface recognition and active tire heating control system of the present invention includes a multi-source information fusion road surface recognition module, a tire heating control strategy module, and an active tire heating control module. Each module realizes data interaction through a CAN bus. The specific structure and function are as follows:

[0052] I. Multi-source information fusion road surface recognition module

[0053] Please see Figure 2 As shown, this module achieves high-precision identification of road surface type and key parameters through multimodal sensor data fusion, and specifically includes 5 units:

[0054] (1) Information input unit

[0055] Integrated with the vehicle's original factory-configured 1080P visible light camera (30fps, 120° field of view) and 16-line LiDAR (detection range). Angular resolution 0.2°), infrared temperature sensor (measuring range) (Sampling rate 10Hz), and simultaneously access wheel speed, acceleration, and steering angle data from the vehicle's CAN bus, forming a multi-source data input of "vision + radar + temperature + vehicle dynamics".

[0056] (2) Information preprocessing unit

[0057] Time synchronization: Based on the timestamp of the LiDAR, the data of the camera (image frame interval 33ms) and the infrared sensor (sampling interval 100ms) are aligned by an interpolation algorithm, with a synchronization accuracy of ≤10ms;

[0058] Denoising: Gaussian filtering (kernel size 3×3) was used to remove salt-and-pepper noise from the camera images, and statistical filtering (number of neighboring points 15, standard deviation multiple 2) was used to remove outliers from the LiDAR point cloud.

[0059] Outlier handling: using The criteria exclude abnormal data from infrared temperature sensors (such as transient jumps). (data), data integrity ;

[0060] Feature standardization: Min-Max normalization is used to map data of different dimensions such as color, texture, and distance to... The interval, the formula is: .

[0061] (3) Feature extraction unit

[0062] A multi-scale feature extraction network is constructed based on a deep learning model: where,

[0063] Shallow network (layers 1-5): Extract road surface color histogram features (3 channels, 64 bins) and texture LBP features (radius 3, neighborhood points 8);

[0064] Middle layer network (layers 6-15): Extracts geometric contour features (road slope, unevenness) of lidar point cloud;

[0065] Deep network (layers 16-20): fuses wheel speed and acceleration data, extracts vehicle driving state related features (such as wheel speed fluctuation coefficient and longitudinal acceleration change rate); the final output is a low-dimensional feature vector with a dimension of 256;

[0066] (4) Model classification and decision-making unit

[0067] An improved SVM classification model is used:

[0068] Model optimization: The penalty parameter C=10 and kernel function parameters of the RBF kernel function were optimized using the PSO algorithm. The classification accuracy rate has been improved to 96.2%;

[0069] Probability output: The SVM output is transformed into a probability distribution for various road surfaces (e.g., 92% for dry land and 8% for wet land) by Platt scaling.

[0070] Decision rule: When the probability is highest When the probability is high, directly output the corresponding road surface type; when the probability is high... At the same time, combined with the steering angle (such as the steering angle) The priority is to determine whether it is a curve or a slope (e.g., slope). The system prioritizes identifying sloping road surfaces for scenario adaptation to ensure decision reliability.

[0071] (5) Recognition result output unit

[0072] The road surface type (e.g., snow accumulation + curves) and the estimated friction coefficient (calculated by fitting the road surface roughness and infrared temperature based on lidar point clouds, with high accuracy) are considered. ), road surface temperature (average data from infrared sensors, accuracy) The data is transmitted to the tire temperature control strategy module via the CAN bus, with a data update frequency of 10Hz.

[0073] II. Tire Temperature Control Strategy Module

[0074] Please see Figures 3-4 As shown, this module generates independent target temperatures for four wheels based on road surface recognition results and vehicle dynamic data, specifically comprising three units:

[0075] (1) Road surface information classification unit: The road surface is divided into two categories through logical judgment:

[0076] Single-state road surface: Contains only one type of road surface (e.g., pure dry ground, pure ice surface), and the determination criterion is the probability of a single type in the multi-source recognition results. ;

[0077] Composite road surface: includes two or more road surface types (such as snow + curve, ice + slope), and the determination condition is that the probability of both types is equal. And there is a steering angle or slope Dynamic characteristics;

[0078] (2) The temperature rise estimation model and arbitration unit includes: information input and calculation subunit, arbitration and decision-making subunit and result output unit;

[0079] The information input and calculation subunit performs target temperature 1 ( During the calculation, six tests were conducted on icy surfaces and four on dry asphalt surfaces over ten cycles. The reference temperature for icy surfaces was 20 degrees Celsius, grade 5, while the reference temperature for dry asphalt surfaces was 10 degrees Celsius, grade 1. The reference temperatures for each road surface can be found in Table 1 below. Calculate the target temperature using the formula. .

[0080] Calculation: Six instances of icy surfaces and four instances of dry asphalt were detected across ten cycles. Icy surfaces: Number of occurrences. reference temperature Security level ;

[0081] Dry asphalt: Number of occurrences reference temperature Security level ;

[0082] ;

[0083] Weight of snow cover: Weight of wetland asphalt: ;

[0084] Substitute into the formula to calculate :

[0085] .

[0086] Road surface type grade Road surface temperature range Reference temperature Dry asphalt 1 10 dry cement 1 10 wetland asphalt 2 12 wetland cement 2 12 Snow-covered road surface 3 14 icy roads 4 16 Icy road surface 5 20

[0087] Table 1 Road Surface Type-Temperature Mapping Table

[0088] The information input and calculation subunit calculates the target temperature 2 ( During calculation, based on the vehicle body Actual vehicle speed on the bus (GPS and wheel speed fusion value), wheel angular velocity Effective rolling radius of the wheel (Preset value 0.32m, error) The formula is ;

[0089] Sensor information confidence level β calculation: Based on 20 samples collected by the infrared temperature sensor within 5 seconds, count the number m of samples below the freezing point (0℃), and combine the signal stability (if the fluctuation of 3 consecutive samples is ≤0.5℃, the stability coefficient is 1, otherwise it is 0.8), the formula is β=(m / 20)×stability coefficient; for example, if 15 of the 20 samples are below 0℃, the stability coefficient is 1, then β=15 / 20×1=0.75;

[0090] Estimate slip ratio And table lookup: using formulas calculate ,For example , ,but ; Query through Table 2 below corresponding ;

[0091] Estimate slip ratio A Target temperature \ 10 12 15 18 20

[0092] Table 2 Slip Ratio-Target Temperature Mapping Table

[0093] (3) Arbitration and Decision-Making Subunit

[0094] Threshold judgment: setting ,like (like Difference If ), then priority arbitration will be applied: icy and snowy roads will be given priority. (Real-time feedback based on road surface type is faster), dry / slippery roads are given priority. (Predictions based on slip ratio are more reliable);

[0095] Dynamic weight allocation: If The first cycle is according to calculate Subsequent period calculations ( (rate of change within 1 second) The weight of the largest absolute value of the growth rate increases by 0.1 (maximum). );For example The temperature increased from 8.6℃ to 9.2℃. ), Increase from 5℃ to 6.5℃ ( ),but Adjusted to 0.6 Adjusted to 0.4 ;

[0096] Temperature protection: If and But currently (like ,current If ), then retain Use this as the current target temperature to avoid excessive temperature fluctuations;

[0097] (4) Result output sub-unit: Outputs the unified target temperature after arbitration. .

[0098] Please see Figure 5 As shown, the complex road surface temperature rise arbitration unit determines a complex road surface by detecting a slope >5° using a slope sensor (accuracy ±0.1°), a steering angle >15° using a steering angle sensor (accuracy ±0.5°), or a difference in friction coefficient between the left and right sides of the road surface >0.1 using a lidar.

[0099] The partition compensation strategy is as follows:

[0100] For a split-type road surface: the friction coefficient of the left side is 0.2, and that of the right side is 0.4 (difference of 0.2). Therefore, the target temperature for the left tire is... Increase by 3℃ from the base (e.g.) Target temperature on the left right side );

[0101] Curving surfaces: Steering angle After calculating the roll moment, the right tire was determined to be the load offset side, and the target temperature increased. (like right side Left side );

[0102] Slope road surface: uphill (slope) The target temperature of the drive wheel (assuming it is the front wheel) increases. (like Front wheel ,rear wheel Downhill (slope) The target temperature of the brake wheel (assuming it is the rear wheel) increases. (like ,rear wheel Front wheel );

[0103] Four-round target temperature output: final generation (Example of split-road surface), data is transmitted to the active tire heating control module via CAN bus;

[0104] , , , Temperature compensation is provided for the open road surface, steering wheel, uphill drive wheel, and downhill brake wheel. The temperature compensation is obtained by referring to tables (Tables 3-6) based on the difference in friction coefficients of the open road surfaces, the steering angle, and the uphill and downhill slopes of the road surfaces.

[0105] Difference in friction coefficient [0.1,0.2) [0.2,0.3) [0.3,0.4) Temperature compensation 3 4 5 8

[0106] Table 3. Difference in Friction Coefficient of Split Road Surfaces - Temperature Compensation Table

[0107] Steering angle [15,25) [25,30) [30,35) 35 Temperature compensation 2 4 5 8

[0108] Table 4 Steering Angle-Temperature Compensation Table

[0109] uphill slope [5,10) [10,15) [15,20) 25 Temperature compensation 3 4 5 8

[0110] Table 5. Uphill Slope-Temperature Compensation Table

[0111] downhill slope [5,10) [10,15) [15,20) 25 Temperature compensation 4 5 6 8

[0112] Table 6 Downhill Slope-Temperature Compensation Table

[0113] III. Active Tire Heating Control Module

[0114] Please see Figure 6 As shown, this module enables precise control of tire temperature and specifically includes four units:

[0115] (1) Temperature acquisition unit: One NTC thermistor (model NCP18WF104F03RC) is embedded in the sidewall of each tire, with a measurement range of Precision Tire temperature is transmitted via a wireless tire pressure monitoring system (TPMS). The data update frequency is 5Hz;

[0116] (2) Temperature control unit: based on the initial tire temperature Ambient temperature Target temperature query: pre-defined "target temperature - - - Drive current mapping table (as shown in Table 7 below) determines the basic drive current. and corresponding PWM duty cycle ( , );For example If the target temperature is 14℃, then ;

[0117] -40 -30 -20 -10 0 10 20 30 -30 22 20 18 16 14 12 10 8 -20 20 18 16 14 12 10 8 6 -10 18 16 14 12 10 8 6 4 0 16 14 12 10 8 6 4 2 10 14 12 10 8 6 4 2 1

[0118] Table 7. Target Temperature 14℃ - Initial Temperature - Ambient Temperature - Drive Current Mapping Table

[0119] Closed-loop control example: using incremental PID algorithm, input deviation Output increment ( , , );For example , , ,but Ultimately, the PWM duty cycle ;

[0120] Current drive unit: Constructed using MOSFET power transistors (model IRF3205) Bridge driver circuit, input PWM signal (frequency) After that, output Continuously adjustable drive current, current ripple It has overcurrent protection ( ), overheating ( ) protection function;

[0121] Built-in heating element in tire: Utilizes a flexible graphene heating film (thickness...) Surface power density The tire is divided into three zones (zone 1 in the center of the tire crown and two zones on both sides of the tire crown) integrated into the tire inner liner. The zones are switched on and off by independent drive circuits. For example, when the left tire on the opposite side of the road needs to be heated quickly, all three zones are heated at the same time. When the temperature is close to the target value, only zone 1 is turned on for heat preservation, which improves energy utilization efficiency by 30%.

[0122] Example 1: Snow accumulation + wet asphalt + split pavement condition

[0123] I. The multi-source information fusion road surface recognition process is as follows:

[0124] 1. Information Input: The camera captures white coverings on the right side of the road and no coverings on the left side; the lidar detects a road surface roughness of 0.8mm on the left and 3.2mm on the right; the infrared sensor detects a road surface temperature of -5℃ on the left and -8℃ on the right; vehicle CAN bus data: vehicle speed 35km / h, wheel speed 22rad / s, steering angle 0°;

[0125] 2. Information preprocessing: After time synchronization, the camera image is denoised, the LiDAR point cloud is filtered, and the temperature and roughness data are standardized to the [0,1] range;

[0126] 3. Feature extraction: The model extracts the "white + high roughness" feature of the right side of the road surface and the "dark + low roughness" feature of the left side;

[0127] 4. Model Classification and Decision: The output shows a 93% probability of wetland on the left and a 95% probability of snow accumulation on the right. The friction coefficients are 0.6 on the left and 0.2 on the right (difference 0.4 > 0.1), classifying it as "snow accumulation + open road surface".

[0128] 5. Output Results: Road Surface Type = Snow Covered, Wet Asphalt + Split Road Surface, Road Surface Temperature The coefficient of friction is 0.4.

[0129] II. Tire temperature control strategies are as follows:

[0130] Road surface information classification: Composite road surface (snow accumulation + wet asphalt);

[0131] Temperature rise estimation models and arbitration:

[0132] Calculation: Snow accumulation occurred 8 times and wet asphalt occurred 2 times in 10 cycles. Snow accumulation: Number of occurrences. reference temperature Security level ;

[0133] Wetland asphalt: Number of occurrences reference temperature Security level ;

[0134] ;

[0135] Weight of snow cover: Weight of wetland asphalt: ;

[0136] Substitute into the formula to calculate :

[0137] .

[0138] Calculation: Vehicle data: Vehicle speed Wheel angular velocity Effective rolling radius of the wheel ;

[0139] Initial slip ratio: ;

[0140] Sensor information confidence level: Of the 20 infrared sensor samples, 14 were below 0℃, with a stability coefficient of 1. ;

[0141] Final estimated slip ratio: ; obtained by looking up the table ;

[0142] arbitration: First cycle ;

[0143] Complex Road Surface Temperature Rise Arbitration: For ordinary composite road surfaces (no split-wheel / turning), the target temperature for all four tires is uniformly 14.35℃. However, if the road surface is judged to be split-wheel, a zone compensation strategy is employed. Split-wheel road surface: The friction coefficient of the left side is 0.6, and the right side is 0.2 (difference 0.4), therefore the target temperature of the right tire is... Increase the temperature by 5°C. Final product .

[0144] III. Active Tire Heating Control:

[0145] Temperature acquisition: ;

[0146] Temperature control (taking the front axle as an example): The target temperature is 14.35℃. (Referring to the table...) ( ); (Taking incremental PID as an example) , Target temperature for the right tire Target temperature 19 The corresponding table lookup result , , , ;

[0147] Current drive: Left output current = Output current on the right side = ;

[0148] Tire heating: 3 zones on the left side are heated simultaneously, and 3 zones on the right side are heated simultaneously; after 3 minutes, the left side... (deviation ), right side (deviation ), enter heat preservation mode (output current drops to 8A);

[0149] Effect verification: After heating, the grip of the left tire increased by 15% and that of the right tire by 40%. The vehicle braking distance was shortened from 55m to 42m, and the lateral offset was reduced from 0.8m to 0.3m, meeting the requirements for safe driving.

[0150] Example 2: Ice surface + turning road condition

[0151] Multi-source information fusion road surface recognition: The camera collects road surface reflections, the lidar detects the road surface roughness of 0.3mm, the infrared sensor detects the temperature of -15℃ and the turning angle of 20°, and the SVM outputs an ice surface probability of 96% and a curve probability of 92%, which is judged as "ice surface + turning road surface".

[0152] Tire temperature control strategy: The right tire is the load offset side, with a target temperature compensation of 5℃. Target temperatures for all four tires: ;

[0153] Active tire heating control: The tire temperature reaches the target value after 4 minutes. When turning, the vehicle's slip angle drops from 6° to 4°, and the steering response delay is shortened from 0.8s to 0.6s, significantly improving handling and stability.

[0154] It is worth noting that in the above system embodiments, the various units are divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of the present invention.

[0155] Furthermore, those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware, and the corresponding program can be stored in a computer-readable storage medium.

[0156] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A multi-source information fusion road surface recognition and active tire heating control system, comprising a multi-source information fusion road surface recognition module, a tire heating control strategy module, and an active tire heating control module connected in sequence by signals, characterized in that: The multi-source information fusion road surface recognition module is used to identify road surface types and evaluate key road surface parameters; The tire heating control strategy module is used to generate independent target temperatures for the four tires. The active tire heating control module is used to regulate the tire temperature to the target temperature.

2. The road surface recognition and active tire heating control system based on multi-source information fusion according to claim 1, characterized in that, The multi-source information fusion road surface recognition module includes an information input unit, an information preprocessing unit, a feature extraction unit, a model classification and decision-making unit, and a recognition result output unit. The information input unit includes a visible light camera, lidar, infrared temperature sensor, and road surface estimation algorithm built into the vehicle, used to collect road surface images, distance, temperature, and vehicle driving-related data. The information preprocessing unit is used to perform time synchronization, Gaussian filtering for noise reduction, and other processes on the raw data from the multi-source sensors. Outlier handling criteria and Min-Max feature standardization are used to form spatiotemporally consistent feature vectors. The feature extraction unit extracts road surface color histogram features, texture LBP features, geometric contour features and vehicle driving state correlation features from the preprocessed data using the ResNet50 deep learning model, and outputs a low-dimensional feature vector. The model classification and decision-making unit adopts an improved support vector machine classification model to process low-dimensional feature vectors and output the probability distribution of dry land, wet land, snow, ice, curves, and split roads. The decision-making module outputs the final road recognition result based on the maximum probability principle and scene adaptation rules. The identification result output unit transmits the road surface type, estimated friction coefficient, and road surface temperature to the tire heating control strategy module.

3. The road surface recognition and active tire heating control system based on multi-source information fusion according to claim 1, characterized in that, The tire temperature rise control strategy module includes a road information classification unit, a temperature rise estimation model and arbitration unit, and a complex road surface temperature rise arbitration unit. The road surface information classification unit divides the road surface into a single state or a composite state through logical judgment. The temperature rise estimation model and arbitration unit include an information input and calculation subunit, an arbitration and decision-making subunit, and a result output subunit. The complex road surface temperature rise arbitration unit determines whether the road surface is complex or sloping by using a road surface slope sensor and a steering angle sensor. When the interpolation of the friction coefficients of the left and right road surfaces is detected At that time, the complex road surface is a split road surface, and the target temperature of the tire on the low friction coefficient side will be increased. When the steering angle The complex road surface is a curved road surface. The roll moment is calculated based on the steering angle, and the target temperature of the tire on the load-off side increases. When the slope The complex road surface is a curved road surface, and the target temperature of the drive wheels increases when going uphill. The target temperature of the brake wheel increases when going downhill. The final output includes four independent target temperatures. The These correspond to the front left, front right, rear left, and rear right tires, respectively; among them, , , , The temperatures are obtained from a table.

4. The road surface recognition and active tire heating control system based on multi-source information fusion according to claim 3, characterized in that, The information input and calculation subunit calculates the target temperature through a dual-path approach. Path 1: Introducing road surface information confidence level ; Value Based on the frequency of occurrence of various road surface results and scene level weighted calculation within 10 recognition periods, In the formula, For the first Number of times this type of road surface occurs For the first Safety level of road surface; The reference temperature corresponding to each road surface condition is queried through the preset road surface type-temperature mapping table, and the target temperature is calculated according to the formula. Path Two: Based on the vehicle's actual speed, wheel angular velocity, and effective wheel rolling radius collected by the vehicle body sensors, the slip ratio formula is used. Calculate the initial slip ratio; Introducing sensor information confidence The slip ratio is estimated by classifying and weighting the number of samples collected from infrared road surface temperature signals, and then calculating the slip ratio using a formula. The final estimated slip ratio is obtained; the target temperature is obtained by querying the preset slip ratio-target temperature mapping table. .

5. The road surface recognition and active tire heating control system based on multi-source information fusion according to claim 3, characterized in that, The arbitration and decision-making subunit acquires the target temperature. Target temperature In addition to tire temperature rise characteristics, a temperature interpolation threshold is set. ;when When, priority arbitration shall be executed; when At that time, the first period is based on weight. Calculate the initial target temperature Subsequent cycles will be based on the temperature growth rate. , Adjust the weights; if and But currently Previous cycle Then keep Use the current target temperature; otherwise, output the normal temperature value.

6. The road surface recognition and active tire heating control system based on multi-source information fusion according to claim 1, characterized in that, The active tire heating control module includes a temperature acquisition unit, a heating control unit, a current drive unit, and a tire-embedded heating element. The heating control unit pre-calibrates the drive current required for the tire to heat up to the target temperature under different initial and ambient temperatures. The corresponding PWM duty cycle is used as feedforward compensation, and the output is dynamically adjusted in combination with closed-loop control. The heating element is controlled by the current drive unit to achieve closed-loop tire heating.

7. A road surface recognition and active tire heating control system based on multi-source information fusion according to claim 2, characterized in that, In the model classification and decision-making unit, the classification model analyzes the feature vectors and outputs the probability distribution of various road surfaces. The decision-making module then synthesizes the probability distribution to obtain the final road surface identification result.

8. The road surface recognition and active tire heating control system based on multi-source information fusion according to claim 6, characterized in that, The heating element is attached to the inner side of the tire tread or integrated into the tire rubber composite material to achieve zoned heating of the tire contact area; the active tire heating control module controls the heating process based on the difference between the actual temperature and the target temperature.

9. A road surface recognition and active tire heating control system based on multi-source information fusion according to claim 1, characterized in that, The system also includes a collaborative pre-aiming unit with the vehicle navigation system and the front vision sensor; the collaborative pre-aiming unit is configured to: identify ahead curves, slopes or potential low-traction road surfaces based on navigation path information and visual pre-aiming information, and instruct the tire heating control module to start or adjust the heating strategy in advance.

10. A road surface recognition and active tire heating control system based on multi-source information fusion according to claim 1, characterized in that, The system's workflow is as follows: Step S1: Continuously collect road surface data and vehicle dynamic data through multi-source sensors; Step S2: Identify the current road surface type and estimate its state parameters based on the collected data; Step S3: Determine the target heating parameters for at least one tire based on the identified road surface type and condition parameters; Step S4: Based on the determined target heating parameters, generate a control signal to the heating actuator; Step S5: Adjust the heating power in real time based on tire temperature feedback.

Images